scenario-kd-scr-ner-half_data-univner_full44
This model is a fine-tuned version of haryoaw/scenario-TCR-NER_data-univner_full on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.0881
- Precision: 0.6065
- Recall: 0.5481
- F1: 0.5758
- Accuracy: 0.9594
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 3e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 44
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 30
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
2.8487 | 0.2910 | 500 | 2.5495 | 0.2969 | 0.0180 | 0.0340 | 0.9243 |
2.158 | 0.5821 | 1000 | 2.1690 | 0.2565 | 0.1146 | 0.1584 | 0.9273 |
1.9133 | 0.8731 | 1500 | 2.1064 | 0.2408 | 0.1652 | 0.1960 | 0.9275 |
1.7757 | 1.1641 | 2000 | 1.9315 | 0.2880 | 0.1886 | 0.2279 | 0.9319 |
1.6744 | 1.4552 | 2500 | 1.8567 | 0.3126 | 0.2055 | 0.2480 | 0.9340 |
1.5971 | 1.7462 | 3000 | 1.8138 | 0.3378 | 0.2627 | 0.2956 | 0.9377 |
1.5019 | 2.0373 | 3500 | 1.7037 | 0.3467 | 0.2767 | 0.3078 | 0.9390 |
1.3519 | 2.3283 | 4000 | 1.6582 | 0.3893 | 0.3186 | 0.3504 | 0.9403 |
1.3481 | 2.6193 | 4500 | 1.6593 | 0.3725 | 0.3350 | 0.3528 | 0.9399 |
1.3011 | 2.9104 | 5000 | 1.5979 | 0.4068 | 0.3300 | 0.3644 | 0.9417 |
1.2064 | 3.2014 | 5500 | 1.5394 | 0.4172 | 0.3637 | 0.3887 | 0.9443 |
1.1639 | 3.4924 | 6000 | 1.5140 | 0.4106 | 0.3982 | 0.4043 | 0.9437 |
1.1121 | 3.7835 | 6500 | 1.4630 | 0.4323 | 0.3763 | 0.4023 | 0.9455 |
1.0805 | 4.0745 | 7000 | 1.4722 | 0.4025 | 0.4082 | 0.4053 | 0.9459 |
1.0035 | 4.3655 | 7500 | 1.4320 | 0.4758 | 0.3955 | 0.4319 | 0.9473 |
0.9826 | 4.6566 | 8000 | 1.4489 | 0.4523 | 0.4275 | 0.4395 | 0.9476 |
0.9617 | 4.9476 | 8500 | 1.3995 | 0.4649 | 0.4294 | 0.4464 | 0.9496 |
0.883 | 5.2386 | 9000 | 1.3944 | 0.4579 | 0.4477 | 0.4528 | 0.9489 |
0.8619 | 5.5297 | 9500 | 1.3635 | 0.4842 | 0.4432 | 0.4628 | 0.9500 |
0.8551 | 5.8207 | 10000 | 1.3569 | 0.4830 | 0.4489 | 0.4653 | 0.9504 |
0.8206 | 6.1118 | 10500 | 1.3573 | 0.4647 | 0.4613 | 0.4630 | 0.9512 |
0.7659 | 6.4028 | 11000 | 1.3415 | 0.4865 | 0.4460 | 0.4653 | 0.9518 |
0.7565 | 6.6938 | 11500 | 1.3367 | 0.4821 | 0.4464 | 0.4636 | 0.9515 |
0.7449 | 6.9849 | 12000 | 1.3177 | 0.4976 | 0.4666 | 0.4816 | 0.9524 |
0.6966 | 7.2759 | 12500 | 1.3129 | 0.5167 | 0.4532 | 0.4829 | 0.9525 |
0.6766 | 7.5669 | 13000 | 1.3272 | 0.4902 | 0.4667 | 0.4782 | 0.9529 |
0.675 | 7.8580 | 13500 | 1.2944 | 0.5046 | 0.4777 | 0.4908 | 0.9529 |
0.6374 | 8.1490 | 14000 | 1.3035 | 0.5464 | 0.4540 | 0.4959 | 0.9535 |
0.617 | 8.4400 | 14500 | 1.2621 | 0.5113 | 0.5136 | 0.5125 | 0.9537 |
0.6172 | 8.7311 | 15000 | 1.2523 | 0.5155 | 0.5158 | 0.5156 | 0.9546 |
0.6023 | 9.0221 | 15500 | 1.2635 | 0.5292 | 0.4992 | 0.5138 | 0.9549 |
0.5671 | 9.3132 | 16000 | 1.2415 | 0.5143 | 0.5223 | 0.5183 | 0.9552 |
0.5602 | 9.6042 | 16500 | 1.2467 | 0.5408 | 0.5005 | 0.5199 | 0.9560 |
0.5565 | 9.8952 | 17000 | 1.2424 | 0.5378 | 0.5092 | 0.5231 | 0.9558 |
0.5292 | 10.1863 | 17500 | 1.2431 | 0.5391 | 0.5184 | 0.5285 | 0.9559 |
0.5152 | 10.4773 | 18000 | 1.2265 | 0.5334 | 0.5377 | 0.5356 | 0.9560 |
0.5217 | 10.7683 | 18500 | 1.2145 | 0.5490 | 0.5168 | 0.5324 | 0.9565 |
0.4982 | 11.0594 | 19000 | 1.2149 | 0.5832 | 0.5064 | 0.5421 | 0.9569 |
0.4801 | 11.3504 | 19500 | 1.2300 | 0.5349 | 0.5190 | 0.5268 | 0.9567 |
0.4723 | 11.6414 | 20000 | 1.2026 | 0.5545 | 0.5237 | 0.5387 | 0.9568 |
0.4768 | 11.9325 | 20500 | 1.2123 | 0.5735 | 0.5165 | 0.5435 | 0.9567 |
0.4573 | 12.2235 | 21000 | 1.1930 | 0.5904 | 0.5116 | 0.5482 | 0.9576 |
0.4447 | 12.5146 | 21500 | 1.2093 | 0.5541 | 0.5419 | 0.5480 | 0.9574 |
0.451 | 12.8056 | 22000 | 1.2137 | 0.5457 | 0.5337 | 0.5396 | 0.9568 |
0.4403 | 13.0966 | 22500 | 1.2029 | 0.5715 | 0.5100 | 0.5390 | 0.9571 |
0.4258 | 13.3877 | 23000 | 1.1842 | 0.5790 | 0.5389 | 0.5582 | 0.9575 |
0.4204 | 13.6787 | 23500 | 1.1901 | 0.5654 | 0.5178 | 0.5406 | 0.9571 |
0.4195 | 13.9697 | 24000 | 1.1973 | 0.5785 | 0.5053 | 0.5394 | 0.9571 |
0.3989 | 14.2608 | 24500 | 1.1863 | 0.5751 | 0.5312 | 0.5523 | 0.9579 |
0.3989 | 14.5518 | 25000 | 1.1764 | 0.5652 | 0.5460 | 0.5554 | 0.9575 |
0.4004 | 14.8428 | 25500 | 1.1896 | 0.6038 | 0.5041 | 0.5495 | 0.9580 |
0.3908 | 15.1339 | 26000 | 1.1819 | 0.5926 | 0.5210 | 0.5545 | 0.9580 |
0.3825 | 15.4249 | 26500 | 1.1591 | 0.5820 | 0.5354 | 0.5578 | 0.9581 |
0.377 | 15.7159 | 27000 | 1.1720 | 0.5723 | 0.5350 | 0.5530 | 0.9575 |
0.374 | 16.0070 | 27500 | 1.1424 | 0.5729 | 0.5406 | 0.5563 | 0.9583 |
0.3591 | 16.2980 | 28000 | 1.1840 | 0.5565 | 0.5498 | 0.5532 | 0.9577 |
0.3608 | 16.5891 | 28500 | 1.1557 | 0.5829 | 0.5431 | 0.5623 | 0.9581 |
0.3636 | 16.8801 | 29000 | 1.1718 | 0.5880 | 0.5275 | 0.5561 | 0.9580 |
0.3547 | 17.1711 | 29500 | 1.1445 | 0.5751 | 0.5497 | 0.5621 | 0.9582 |
0.3468 | 17.4622 | 30000 | 1.1362 | 0.5938 | 0.5324 | 0.5614 | 0.9588 |
0.3444 | 17.7532 | 30500 | 1.1412 | 0.5984 | 0.5470 | 0.5715 | 0.9590 |
0.3379 | 18.0442 | 31000 | 1.1374 | 0.5836 | 0.5445 | 0.5634 | 0.9585 |
0.3334 | 18.3353 | 31500 | 1.1453 | 0.5808 | 0.5292 | 0.5538 | 0.9586 |
0.3335 | 18.6263 | 32000 | 1.1363 | 0.5843 | 0.5485 | 0.5659 | 0.9588 |
0.33 | 18.9173 | 32500 | 1.1517 | 0.5939 | 0.5451 | 0.5685 | 0.9588 |
0.3262 | 19.2084 | 33000 | 1.1323 | 0.6000 | 0.5278 | 0.5616 | 0.9585 |
0.3233 | 19.4994 | 33500 | 1.1322 | 0.5906 | 0.5380 | 0.5631 | 0.9587 |
0.3188 | 19.7905 | 34000 | 1.1306 | 0.5869 | 0.5527 | 0.5693 | 0.9587 |
0.3164 | 20.0815 | 34500 | 1.1411 | 0.5946 | 0.5296 | 0.5602 | 0.9585 |
0.3137 | 20.3725 | 35000 | 1.1301 | 0.5960 | 0.5428 | 0.5681 | 0.9587 |
0.3085 | 20.6636 | 35500 | 1.1298 | 0.5890 | 0.5337 | 0.5600 | 0.9586 |
0.3105 | 20.9546 | 36000 | 1.1364 | 0.5733 | 0.5442 | 0.5584 | 0.9588 |
0.3047 | 21.2456 | 36500 | 1.1228 | 0.6020 | 0.5457 | 0.5725 | 0.9594 |
0.2971 | 21.5367 | 37000 | 1.1253 | 0.6087 | 0.5380 | 0.5712 | 0.9591 |
0.303 | 21.8277 | 37500 | 1.1135 | 0.6096 | 0.5460 | 0.5760 | 0.9592 |
0.3018 | 22.1187 | 38000 | 1.1254 | 0.6101 | 0.5449 | 0.5757 | 0.9590 |
0.2906 | 22.4098 | 38500 | 1.1173 | 0.5975 | 0.5543 | 0.5751 | 0.9592 |
0.2931 | 22.7008 | 39000 | 1.1113 | 0.5997 | 0.5432 | 0.5701 | 0.9592 |
0.2934 | 22.9919 | 39500 | 1.1233 | 0.6080 | 0.5428 | 0.5736 | 0.9591 |
0.2885 | 23.2829 | 40000 | 1.1213 | 0.6074 | 0.5480 | 0.5762 | 0.9590 |
0.2839 | 23.5739 | 40500 | 1.1122 | 0.6030 | 0.5403 | 0.5699 | 0.9590 |
0.2892 | 23.8650 | 41000 | 1.1152 | 0.5958 | 0.5396 | 0.5663 | 0.9584 |
0.2836 | 24.1560 | 41500 | 1.1141 | 0.6037 | 0.5467 | 0.5738 | 0.9593 |
0.2801 | 24.4470 | 42000 | 1.1061 | 0.5908 | 0.5519 | 0.5707 | 0.9592 |
0.2817 | 24.7381 | 42500 | 1.1044 | 0.6056 | 0.5503 | 0.5766 | 0.9591 |
0.2814 | 25.0291 | 43000 | 1.1027 | 0.6142 | 0.5491 | 0.5798 | 0.9597 |
0.2775 | 25.3201 | 43500 | 1.1090 | 0.6068 | 0.5438 | 0.5736 | 0.9591 |
0.2778 | 25.6112 | 44000 | 1.1048 | 0.6068 | 0.5491 | 0.5765 | 0.9597 |
0.2708 | 25.9022 | 44500 | 1.1111 | 0.6030 | 0.5467 | 0.5734 | 0.9590 |
0.2755 | 26.1932 | 45000 | 1.1018 | 0.6089 | 0.5481 | 0.5769 | 0.9595 |
0.2698 | 26.4843 | 45500 | 1.1116 | 0.6023 | 0.5305 | 0.5641 | 0.9588 |
0.2699 | 26.7753 | 46000 | 1.0993 | 0.6102 | 0.5421 | 0.5741 | 0.9595 |
0.2717 | 27.0664 | 46500 | 1.0905 | 0.6031 | 0.5425 | 0.5712 | 0.9590 |
0.2657 | 27.3574 | 47000 | 1.0948 | 0.6024 | 0.5586 | 0.5797 | 0.9598 |
0.2675 | 27.6484 | 47500 | 1.0910 | 0.6159 | 0.5501 | 0.5812 | 0.9596 |
0.2676 | 27.9395 | 48000 | 1.0930 | 0.6018 | 0.5484 | 0.5739 | 0.9596 |
0.2652 | 28.2305 | 48500 | 1.0991 | 0.6024 | 0.5480 | 0.5739 | 0.9591 |
0.2637 | 28.5215 | 49000 | 1.0981 | 0.6058 | 0.5468 | 0.5748 | 0.9594 |
0.2656 | 28.8126 | 49500 | 1.0988 | 0.6060 | 0.5397 | 0.5710 | 0.9594 |
0.2628 | 29.1036 | 50000 | 1.0986 | 0.6094 | 0.5510 | 0.5787 | 0.9597 |
0.2622 | 29.3946 | 50500 | 1.0884 | 0.6079 | 0.5465 | 0.5756 | 0.9598 |
0.2602 | 29.6857 | 51000 | 1.0995 | 0.6065 | 0.5400 | 0.5714 | 0.9594 |
0.2646 | 29.9767 | 51500 | 1.0881 | 0.6065 | 0.5481 | 0.5758 | 0.9594 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.1.1+cu121
- Datasets 2.14.5
- Tokenizers 0.19.1
- Downloads last month
- 4
Model tree for haryoaw/scenario-kd-scr-ner-half_data-univner_full44
Base model
FacebookAI/xlm-roberta-base
Finetuned
this model